geotechnical monitoring for safe excavation of large · pdf filegeotechnical monitoring for...
TRANSCRIPT
Geotechnical Monitoring for Safe Excavation of Large Rock Cavern: A Case
Study
A.Mandala, C. Kumar
b, A. Usmani
b and A. Nanda
b
aDepartment of Civil Engineering, VNIT-Nagpur, India
bEngineers India Ltd., New Delhi, India
E-mail: [email protected], [email protected] , [email protected], [email protected].
Abstract - The present study highlight the methods used to
monitor the geotechnical behaviour and
performance of the cavern during excavation in
order to verify and, if necessary, adjust the rock
support to ensure safe construction at every stage.
The present geotechnical monitoring practice for
underground structures involve convergence
monitoring with the help of optical targets and
rock mass displacement with help of bore hole
extensometers. More importantly it involve
detailed planning to finalize the position of each
monitoring instruments based on location and
orientation of geological features. Further it is
equally important to ensure proper recording of
monitoring data in order to analyze and take
immediate action in case of any adverse situation.
This requires a dedicated high end automated
software, which can record, analyse and produce
significant results out of large quantity of
recorded data which otherwise turns out to be
only ravage. Initially geotechnical monitoring was
carried out in recently excavated zone of cavern
on daily basis. Further based on continuous
monitoring data for at least one week, frequency
of monitoring is further decided. In most of the
cases the deformation of rock mass was quite less
compare to the alarming values which were
evaluated based on detailed design for different
rock classes.
Keywords -
Underground excavation, Cavern, Tunnel, Geotechnical
Monitoring
1 Introduction
This paper discusses in detail geotechnical
monitoring activity carried out during construction of
an underground unlined rock cavern complex for
storage of crude oil at western part of India. The total
storage capacity was 2.5MMT of crude oil. As four
numbers of rock cavern units were to be built, for the
ease of construction, the caverns were divided into
two parts. Each part consists of two U shaped storage
units with two legs. Each leg was extended almost
700 m, thus total length of cavern was around six
kilometres. This project involve the excavation of
vertical shaft for crude intake and out let, access
tunnel to facilitate excavation at required depth, water
curtain tunnel for continuous ground water recharge
and the large cavern for storage space. However,
present study only concentrates on the geotechnical
monitoring activity implemented inside the cavern as
it was most critical due to its large size and extent.
Each U shaped cavern unit (in plan) is designed with
30.0 m high and 20.0 m wide (in cross section) to
create the required storage space. On and average
cavern crown was 60.0 m below the ground surface
with a trend of 600/240
0.
As this construction involves lot of excavation
activity at different depth and direction, it was very
essential to ensure global stability of this zone along
with the stability of each structure. Any underground
excavation involves certain amount of geological
uncertainties. The more the extent of such excavation
more the uncertainties have to deal with. So, the
available option to handle such situation was
continuous geotechnical monitoring, Geological
assessment and suitable design provision to
incorporate different geological conditions during
excavation.
The support philosophy of all underground
openings within the project is based on staged
excavation, incremental installation of rock support
measures and verification by monitoring. In addition
robust design approached was followed to
The 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC 2014)
accommodate modification of rock supports based on
actual characteristic of rock mass, thus leading to cost
effective and practical rock support.
2 Geological Assessment of Site
Geological and geotechnical investigation was
carried out in three phases during the period 2005 to
2010. In total 20 nos of bore holes were drilled in
vertical and some of them in inclined orientation in
each phases. Initially six numbers of boreholes were
drilled in the year 2005 at some specific locations
based on the information available from geophysical
survey as well as out crops observed during site visit.
The data obtained from first phase survey was used to
develop tentative Geological map of this site
projecting the extension and orientation of
different features. Based on this map, second phase of
investigation was planned which includes another six
bore holes to ensure the possibility of projected
geological features. Finally detailed geological map
was developed after completion of third phase
investigation which includes additional drilling of
eight numbers of bore holes. This geological map
kept on updating as excavation progress
incorporating the geological information obtained
from excavated zone. Further this map was used to
plan the location of optical targets and bore hole
extensometer along cavern section for continuous
monitoring of critical area.
2.1 Geology
This area was located within ’Peninsular
Gneisses’ (M.S. Krishnan, (1953). The Peninsular
Gneisses are characterized by heterogeneous mixture
of different types of granite intrusive into the
schistose rocks after the latter were folded and
crumpled. Within the project area, Peninsular Gneiss
is by far the most represented rock type. However,
types of granite are also found in some places. The
details of the structural discontinuities within the rock
mass as determined from the mapping reveal 3
significant joint sets, 1 minor joint set and the
foliation as show in Table 1.
Generally it was expected that very poor to poor
quality rock will be encountered up to a depth of
approximately 3 to 5 m below ground level, whilst
fair and good rock is anticipated greater than 20 to 30
m below ground level. The local geological
conditions of the project site are characterized by
three geotechnical relevant rock mass layers of
varying thickness.
Table 1: Observed Geological features
Features Dip Direction Dip.
Joint set 1 070 85
Joint set 2 100 85
Joint set 3 200 85
Minor Joint set 045 45
Foliation 020 85
The top layer is varying in thickness and consists of
residual soil and lateritic. Immediate second layer is
characterized by weather, jointed and fractured rock
with boulders with varying thickness and bottom
layer is hard fresh and massive rock. Most of the
project component is located in the bottom layer (bed
rock) except initial portion of access tunnel and shaft.
No major tectonic event has occurred in this zone
subsequent to formation of these rocks.
2.2 Geotechnical Assessment
The main geotechnical failure modes expected in
the caverns of this storage complex are sliding of
blocks/wedges along discontinuities, toppling and
rotational failures and the development of overbreak
in and around zones of fractured and weather rock.
Weathering and ground water are deemed to have had
little influence on the regional rock mass condition.
The main influencing factor is the quantity and
quality of discontinuities. A number of small fault
like features were recorded during the investigation.
However, it was expected to have little impact of
these faults on rock mass behaviour in most cases. No
major pre existing inactive and active mass
movements were detected. However, this area is
classified as Zone-III with seismic coefficient of
about 0.04g as per Seismic Zone map of India (IS
1893).
Relatively high horizontal stress (SH) in the range
of app. 6 to 9 MPa was envisaged along the cavern
roof. For the typical vertical stress at the site of about
4 MPa, corresponding approximately 50 to 80m of
overburden, the typical values of the in-situ stress
ratios was in a range of 2.2 MPa.
3 Geotechnical Monitoring Plan
In this project two differ types of instrumentations
were used to measure the deformation i.e.
POSTER PAPER
convergence monitoring by optical targets and rock
mass deformation by extensometer. Convergence
monitoring was carried out along the cavern section.
However, deformation measurement by extensometer
was only limited to geological hot spot as observed
during investigation and subsequently substantiated
during the excavation of other components (access
tunnel and water curtain tunnel) of this project,
specifically from the excavation of water curtain
tunnel which was located just 20.0 m above the
cavern crown.
3.1 Optical Targets
The deformation monitoring of excavation
surfaces of cavern using 3D measurement of optical
targets (in x-y and z direction) was the main
geotechnical monitoring method in order to assess the
rock mass behaviour during excavation.
Optical targets were fixed to reference points
through bolting to the rock surface. The bolt was
installed through a drill hole of diameter 25 mm upto
a depth of 220 mm from rock surface. Then bi reflex
target was fitted to the installed bolt with the help of
the screw. Monitoring section within the caverns are
typically arranged at every 25.0 m of interval with
five targets within the top heading (one in crown and
two at each side) and 4 targets along each side wall as
shown in Fig. 1. However, additional section of
optical targets was installed based on observed
geological features in updated geological map.
Fig. 1: Cavern cross section with optical targets and
excavation stage.
The dotted line in the cross section of cavern
shows the excavation stage of each bench. Initially
pilot tunnel was excavated which was further
extended by side slashing to complete the heading
excavation. Similarly each bench excavation is also
associated with side slashing to complete each bench
thus total three bench excavations were carried out.
All three optical targets at the crown location were
installed immediately after the completion of
necessary support activity in pilot tunnel and other
two targets at heading were installed after completion
of first bench excavation. The targets at the wall were
installed after completion of subsequent bench
excavation. Initially at few locations, the damage of
optical targets was reported during first bench
excavation. Therefore, necessary precaution was
taken to protect the targets by removing it from the
bolt during blasting activity and placed thereafter.
3.2 Extensometer
Extensometer was installed in drilled boreholes
from the invert of water curtain tunnel at some
geologically critical locations as identified from
updated geological map. This water curtain tunnel is
located just 20.0 m above the crown of the cavern
which will be filled with water during operation to
ensure continuous flow of water around caverns to
maintain require hydrostatic pressure around the
cavern for confinement of crude oil. This tunnel is
extended along the cavern length.
At each location two different extensometers were
installed toward the cavern on either side of the water
curtain tunnel as shown in Fig. 2. Extensometer was
installed in drilled boreholes with three point
instrument at 5.0 m interval thus maximum length of
20.0 m and furthest point is approximately 5.0 m
away from cavern roof. This will monitor the rock
mass movement at crown of the cavern which was
envisaged as most critical part from stability point of
view.
Each extensometer rod was grouted where the rod
end is located in rock to develop strong bonding
between the rod and surrounding rock mass. Thus
minimize the chances of slippage between
extensometer rod and surrounding rock. The head of
the instrument was housed within a steel protection
cap to avoid disturbance and damage from different
construction activity. The sole purpose of these
extensometers was to monitor the rock mass
behaviour during construction.
Once the cavern was fully excavated and its
stability was further ascertained through the
stabilized extensometer monitoring data, the
instrument was abandoned and all void within the
instrument was grouted with cement grout to avoid
any potential passageway of water towards the cavern
crown.
It is usually works based on the assumption that
the deepest anchor is in stable ground and any change
in anchor spacing is interpreted as rock mass
The 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC 2014)
movement. Here, the topmost anchor point (close to
water curtain tunnel) was considered as in stable
ground and will indicate any relative movement of
rock in cavern crown during cavern excavation which
will be recorded in it’s head assembly.
Fig. 2: Location and extent of Extensometer
from Water Curtain Tunnel
4 Geotechnical Monitoring Procedure
Monitoring team comprises Project manager,
Shift Manager, Site Engineer and Surveyor worked
till the end of construction which completed in the
year 2013 involving the underground excavation of
34.9 Lakh cum rock without any accident related to
stability of rock.
4.1 Convergence Reading
Reading from optical targets was taken with the
help of total station survey. The control points were
fixed with reference to the original project co-
ordinates at every section where the targets were
fixed. First set of readings shall be zero readings of
the targets. Data of all the readings were stored in the
memory module of total station automatically and
then down loaded it to the computer for further
processing.
4.2 Extensometer Reading
Grouted anchored type borehole extensometer
composed of head assembly, vibrating wire
displacement sensors, connecting rod assembly and
anchor or reinforcement bar were installed. Extension
rods were installed in to the bore hole up to the
required depths. Then the extensometer sensors were
fixed after four days of grouting and initial reading
were taken from read out unit. Data were stored and
plotted to find the trend of rock mass behaviour.
4.3 Monitoring Frequency
The data manually obtained from optical targets
and extensometers were entered and saved
electronically in spreadsheet and plotted with time
(frequency).
Table 2: Monitoring frequencies for convergence
monitoring by Optical targets.
Instrument Frequency Condition-1 Condition-2
Optical
Targets
Once per
day
Within
30.0m from
the
excavation
face
The
differential
deformation
between the
current and
the previous
reading is
more than
5.0 mm.
every other
day
within 30.0m
from the
excavation
face
The
differential
deformation
between the
current and
the previous
reading is
more than
2.0 mm.
Once every
week
between
30.0m and
60.0m from
the
excavation
face
The
differential
deformation
between the
current and
the previous
reading is
larger than
monitoring
accuracy
once per
month
excavation
face is more
than 60.0 m
away
The
differential
deformation
between the
current and
the previous
reading is
within
monitoring
accuracy
POSTER PAPER
These frequencies were mentioned in Table 2 & 3
with different condition. However, this frequency
was further modified based on actual behaviour of
rock mass observed after plotting and interpreting the
recorded data. Observed data as per given frequency
was plotted and uploaded to the centralize server and
all concern person could have the access to the server
whenever required. This was followed by the
interpretation of all monitoring data by design
engineer to assess the overall stability at certain stage
to decide and verify further excavation strategy.
Table 3: Monitoring frequencies for Extensometers.
Instrument Frequency Condition-1 Condition-2
Extensometer
Once per
day
Excavation
face in
cavern
gallery is
within 40.0
m of the
instrument
The
differential
deformation
between the
current and
the previous
reading is
still more
than 5.0 mm
Once
every
week
Excavation
face in
cavern
gallery is
more than
40.0 m away
from the
instrument.
The
differential
deformation
between the
current and
the previous
reading is
larger than
monitoring
accuracy.
Once per
month
Excavation
face in
cavern
gallery is
more than
40 m away
from the
instrument.
The
differential
deformation
between the
current and
the previous
reading is
within
monitoring
accuracy.
5 Monitoring Results and Discussion
In such large projects, compilation and recording
of monitoring data is an important aspect and cannot
be left to simple practice of sheets and tables.
Presently with the advent of technology, this filed of
data management has also witnessed lot of
upgradation and automation. Therefore in this project
also, all monitoring data were managed through a
dedicated geo monitoring software, which works as a
tunnel information system for continuous monitoring
by concerned person at site as well as design office.
Deformation value as observed in optical targets was
plotted in all three direction (two horizontal and
vertical) with respect to time of observation.
Similarly, extensometer reading of three anchor
points was also plotted with respect to time of
observation. All these results were compared
regularly in reference with different limit value.
5.1 Deformation Limit
Monitoring values were used to evaluate the rock
mass condition by comparing measurement results
against expected soil/rock deformation. Different
monitoring limit values were introduced to categorise
measurements in terms of their severity. This limit
was developed from the expected deformation in
detailed design calculation and functioning and
maintenance of the structure. In the detailed design
calculation, rock mass as observed in this area was
classified broadly in three classes as per their Q
values i.e. Class 1. Good and Very Good Rock, 2.
Fair Rock and 3.Poor Rock and expected deformation
were defined for each of them. For all these three
classes two limits were fixed namely “trigger level”
and “allowable level”. Trigger level is defined as the
permitted maximum displacements of a particular
structure or area, which is critical to the safety,
functioning of the structure. It was considered as 80%
of design value.
Table 4: Monitoring limit values for different rock
Class
Rock Class
(Q-value)
Top Heading Benching Trigger
(mm)
Allowable
(mm)
Trigger
(mm)
Allow
able
(mm)
Very Good
and Good 5.0 10.0 5.0 8.0
Fair 5.0 10.0 5.0 7.0
Poor 24.0 42.0 13.0 22.0
Whereas allowable level, is the maximum limit which
should not exceed in any case and this was set at
140% of the design value. Absolute values as
considered in this project for different rock class at
various stages of excavation are reported in Table 4.
The 31st International Symposium on Automation and Robotics in Construction and Mining (ISARC 2014)
5.2 Interpretation
In most of the case, the observed deformation in
geotechnical monitoring data observed from optical
targets was well within the trigger limit. There was
not a single incident reported where the deformation
exceed the allowable limit. However, at few locations
sudden increase in deformation was observed due to
the major excavation activity in that vicinity. When
the deformation value exceed the trigger value at
some section of the cavern, that location was visually
inspected by engineer to check the possible distress in
terms of increased seepage or any crack in the
installed shotcrete to allow further excavation.
Fig. 3: Deformation plot for vertical and
horizontal direction upto completion of project
Typical deformation of cavern section plotted in Fig.
3 for poor rock condition where relatively higher
deformation was observed.
Fig. 4: Extensometer reading at poor rock
condition
Similar trend of deformation was also observed in
Extensometer reading as presented in Fig. 4. In this
case increase in the deformation was observed during
the pilot tunnel and side drift at heading excavation.
Afterward there was no significant increase in
deformation due to subsequent bench excavation. In
both the measurement, the deformation values in poor
rock condition were remarkably low compare to the
trigger value at different stage of excavation. This
may be due to the conservative design approach as
followed in this project for poor rock class. However,
the incident of poor rock class encountered was less
than 5% as compare to total excavation length.
6 Observation and Comments
This paper was specifically developed to highlight
the systematic approach followed for large cavern
excavation to minimize the risk of rock fall due to
sudden change in geology. Initially, three stage
geological investigations were quite efficient to
understand the rock mass character more accurately.
Particularly it helped to minimize the occasion of
geological surprises involved in underground
excavation. Moreover, continuous design updation
activity employing high end dedicated software’s
based on latest geological information along with
geotechnical monitoring helps to use rock support
system more efficiently and economically. However,
geotechnical monitoring method using optical targets
need to be improved for such large excavation. It was
observed in many occasion during the excavation that
the total station survey to take the reading suffer from
accuracy due to disturbance from construction
activity. The initial deformation which may have
occurred immediately after excavation can’t be
recorded in optical targets measurement. This need to
be addressed in future study by cost effective
automatic instrumentation. Thus monitoring data will
be more accurate and will definitely minimize the
scope of human error involved in recording and
uploading such huge data. Also the application of
mobile technology can help to keep alert all
concerned round the clock.
References
[1] Design Package for Cavern Storage
4923/REP/PUA/0310, Engineers India Ltd., New
Delhi, India.
[2] Geotechnical Monitoring and Instrumentation,
4923/DSB/PUA/0014 Engineers India Ltd., New
Delhi, India.
[3] Krishnan, M. S. The structural and tectonic
histo-y of Jndia ~Mem, Geol, Surv, India,
vol.81. p.1-93.1953.
[4] I.S. 1893 part-1; Indian Standard Code for
Earthquake Intensity.
POSTER PAPER